Veterinary Immunology and Immunopathology, 34 ( 1992 ) 47-61 Elsevier Science Publishers B.V., Amsterdam
47
Purification and characterization of porcine C3. Studies of the biologically active protein and its split products K a r l - E r i k S t o r m a, G 6 s t a A r t u r s o n b a n d U l f R . N i l s s o n a aDepartment of Clinical Immunology and Transfusion Medicine, UniversityHospital S- 751 85 Uppsala, Sweden bBurn Center, UniversityHospital S-751 85 Uppsala, Sweden (Accepted 28 November 1991 )
ABSTRACT Storm, K.-E., Arturson, G. and Nilsson, U.R., 1992. Purification and characterization of porcine C3. Studies of the biologically active protein and its split products. Vet. Immunol. Immunopathol., 34: 47-61. Separation techniques for obtaining pure and biologically active swine C3 have been improved in this study. Using these procedures and through the further characterization of porcine C3, the possibilities for developing more specific techniques for the analysis of the complement system in swine have been improved. Plasma was initially treated with protease inhibitors, polyethylene glycol (PEG)fractionation, plasminogen-depletionand a rapid chromatographic desalting step. The essential fractionation was carried out by DEAE-Sephacel chromatography. Contaminants were removed by sizeexclusion ( Sepharose CL-6B )- and hydroxylapatite-chromatography. The final recovery reached 56% with 73% retaining specific hemolytic activity. The amino acid composition (98.33%), the functional compatibility and the secondary structure of fragments and intact protein indicate a high degree of homology with human C3. In contrast with the findings of earlier studies was the considerable immunologic cross-reactivity observed with human C3, and the size difference between the human and the swine C3-fl subunit, which was found to be 10 kDa lighter than the human analogue. The finding that the swine C3b/iC3b/C3c fragments do not separate from C3 by agarose electrophoresis, unlike the human analogues, demonstrated that this commonly used simple parameter for the detection of complement activation cannot be used in the porcine model. ABBREVIATION EA, sheep erythrocytes sensitized with rabbit hemolysin; EAC4°Xy2, the stable cell intermediate made up of EA and the complement components C4 and C2 (iodine-oxidized), forming the classical third component of complement (C3) convertase; PEG, polyethylene glycol; SDS, sodium dodecyl sulphate; SDS-PAGE, sodium dodecyl sulphate-polyacrylamide gel electrophoresis.
Correspondence to: Karl-Erik Storm, Department of Clinical Immunology and Transfusion Medicine, University Hospital, S-751 85 Uppsala, Sweden.
© 1992 Elsevier Science Publishers B.V. All rights reserved 0165-2427/92/$05.00
48
K.-E. STORM ET AL.
INTRODUCTION
Complement is an important effector system which is involved in the host defense and in numerous vital reactions of the inflammatory process, e.g. vascular permeability, opsonization of foreign cells and proteins and regulation of the immune response. It comprises more than 30 plasma and membrane proteins, which are organized in two activation pathways, one predominantly dependent, the other independent of antibodies, and which both trigger the various biologic effects of the common terminal reaction sequence (MiillerEberhard, 1988). C3 is the most abundant complement protein and participates in both activation pathways. Specific interactions with cell surface molecules, many serum proteins and proteins of foreign origin indicate the presence on C3 of at least ten specific binding sites. The fragmentation during activation modulates the expression of these binding domains, which are further tuneable both by the action of co-factors and by the covalent linkage of the molecule by its reactive internal thiolester (Lambris, 1988 ). Several new techniques have been designed for the analysis of the human complement system, and by the assessment of its physicochemical alterations, activation can be quantified precisely (Cooper, 1988 ). These new techniques therefore increase the possibilities of evaluating the relative importance of complement compared t o other effector systems such as the coagulation-fibrinolytic andkinin-gener~iting systems (Sundsmo and Fair, 1985). The techniques for analysis of the porcine complement system are far less developed as reflected in recent studies of inflammation which limit the complement analysis to the hemolytic test of the entire complement sequence. New analytical procedures have to be developed to make this model available for complement analysis at a level comparable to that possible for the human. An essential step is to make isolated components and fragments of the swine complement system available for specific analysis. Although several porcine complement proteins have been isolated, little effort has been made to study their reaction in the homologous system (Linscott, 1986). In view of the lack of information on the specific activity of purified swine C3 in earlier publications (P~tques, 1980; Zidkov~i et al., 1984), it seemed justified to apply more efficient techniques (Nilsson et al., 1972; Hammer et al., 1981 ) to the isolation, identification and partial characterization of biologically active porcine C3 in the present report. MATERIALS AND METHODS
Approximately 15-week-old pigs were exsanguinated under general anaesthesia with Ketamin as described (Borg et al., 1985 ). The blood was collected
PURIFICATION AND CHARACTERIZATIONOF PORCINE C3
49
in plastic bags containing anti-coagulant (CPD-Sagman, Terumo, Pharmacia, Uppsala). After separation of cells by centrifugation the plasma was kept frozen at - 70 ° C.
C3 separation procedure The sequential steps of the separation procedure can be followed in the left column of Table 1. Swine plasma was treated with inhibitors and polyethylene glycol (PEG) in accordance with the initial anti-proteolytic procedure described by Hammer et al. ( 1981 ). The soluble fraction was applied to a Lysine-Sepharose R (Pharmacia, Uppsala) column ( 5 × 5.5 cm, 195 ml h - 1). The bulk of eluted protein was pooled and transferred by gel filtration on Sephadex R G-25 (Pharmacia, Uppsala) ( 10 × 28.8 cm, 2647 ml h - ~) to buffer A and applied to the DEAE-Sephacel R (Pharmacia, Uppsala) column (Fig. 1 ). Eluted fractions containing C3 were identified by rocket electrophoresis and by hemolytic C3 screening (Figs. 1 and 2). Appropriate fractions were pooled and concentrated by PEG-precipitation according to Hammer et al. (1981). The precipitate was dissolved in approximately 20 ml equilibration buffer containing 0.05% Tween-20, and separated by Sepharose R CL-6B (Pharmacia, Uppsala) chromatography (Fig. 3). The eluate was analysed hemolytically and by fused rocket electophoresis against anti-swine C3 and anti-swine serum proteins respectively. C3 containing fractions were pooled and concentrated. The resulting precipitate of the Sepharose CL-6B pool was dissolved in approximately 20 ml buffer A (Fig. 4 ) containing 0.05% Tween 20, and applied to a column of hydroxylapatite (DNA-grade, Bio-Rad, Richmond, CA) under conditions modified (Fig. 4) from Nilsson et al. (1972). After similar TABLE 1 Summary of yields Fraction
Volume Total protein pool (ml) Conc. Recovery (mgm1-1) (%)
Plasma 360 PEG-supernatant 400 Lysine-Sepharose 460 Sephadex G-25 1010 DEAE-Sephacel 378 Sepharose CL-6B 71.5 Hydroxylapatite 119
37 34 29 12 1.4 3.4 1.5
100 101 99 91 3.9 1.8 1.3
Immunochemical activity
C3 activity
Conc. (mg ml -I )
Recovery Conc. Specific Recovery (%) ( z m l -~) (z [C3] -~) (%)
0.99 0.84 0.76 0.33 0.79 3.3 1.7
100 94 98 92 83 66 56
296 241 169 109 319 1548 365
298 287 223 333 405 471 218
100 90 73 102 113 104 41
50
K,-E. STORM ET AL.
% z
A 28°
mS
so ~
13 wash
i i ........
1
0 1 gradient
2
i
,o
3
v o l u m e (I)
Fig. 1. Anion exchange chromatography of Sephadex G-25 pool. Conditions: column, DEAESephacel (5 × 51.4 cm); sample size 1010 ml; buffer A, 3.2 mM K/Na/H/PO4, 6.4 mM EDTA, 6.4 mM benzamidine-HCl, 31.8 mM EACA, adjusted to 3.0 mS + 4 ° C with NaC1, pH 7.4; buffer B, buffer A adjusted to 18.0 mS+4°C with NaC1; gradient, 620 ml buffer A, 3000 ml 0-100% B, 1000 ml buffer B; flow rate 120 ml h- ~;fraction volume, 10 ml; pooled material, fraction 188 through 225; detection, absorbance 280 nm ( ), conductivity (mS) at + 4°C ( ..... ), relative C3-hemolytic activity (- - -), 100%=maximum z.
anti-C5
anti-C3
............
anti-C4
000000000000000000000000 0000000000000000000000000 I
1000
I
I
2000
I
I
3000
I
v o l u m e (ml)
Fig. 2. Fused rocket immunoelectrophoresis of eluted fractions from DEAE-Sephacel. C3 is detected with specific anti-swine C3. Cross-reacting anti-human antisera are used for the C4 and C5 detection.
screening as for the eluate o f the p r e c e d i n g c o l u m n , the p e a k was p o o l e d a n d concentrated. H u m a n C3 was p r e p a r e d using the s a m e p r o c e d u r e s with the e x c e p t i o n t h a t the a d d i t i o n o f T w e e n 20 in the s o l u b i l i z a t i o n o f P E G - p r e c i p i t a t e s was omitted.
PURIFICATION AND CHARACTERIZATION OF PORCINE C3
0
260
460
660
51
volume (ml)
Fig. 3. Gel filtration of the DEAE-Sephacel pool. Conditions: column, Sepharose CL-6B (2.6 × 95 cm); sample size 21.5 ml; buffer, 100 mM K/Na/H/PO4, 150 mM NaC1, 5 mM EDTA, 50 mM EACA, pH 7.4; flow rate 20 ml h- ~h; fraction volume, 5 ml; pooled material, fraction 61 through 75; detection, absorbance 280 nm ( ), C3-hemolytic activity in percent z-units of highest value (- - -). -,o A 280 100 3
50
mS f 20
21
wash
i ~ i ....... ....
0 gradient
500
10
1000 volume (roll
Fig. 4. Hydroxylapatite chromatography of Sepharose CL-6B pool. Conditions: DNA-grade hydroxylapatite (5 × 10 cm); sample size 18.8 ml; buffer A, 20 mM K/Na/H/PO4, 100 mM EACA, adjusted with buffer C to 7.5 mS+4°C, pH 7.9; buffer B, 20 mM K/Na/H/PO4, 100 mM EACA, adjusted with buffer C to 20.0 mS + 4 ° C, pH 7.9; buffer C, 650 mM K/Na/H/PO4, 100 mM EACA, pH 7.9; gradient, 300 ml buffer A, 1000 ml 0-100% B; flow rate 30 ml h-~; fraction volume 5 ml; pooled material, fraction 81 through 105; detection, absorbance 280 nm ( ), conductivity (mS) at + 4 °C ( ..... ), C3-hemolytic activity in percent z-units of highest value (- - -).
C3b-binding and hemolytic techniques C o n v e r t a s e - d e p e n d e n t C 3 b - b i n d i n g was assessed with 125I-labelled C3. Labelling o f C3 with 125I ( A m e r s h a m Sweden, S o l n a ) to a final activity o f 1.5 × 106 c p m / t g - 1 was p e r f o r m e d a c c o r d i n g to N i l s s o n a n d Nilsson ( 1985 ). EA a n d c e l l - b o u n d c o n v e r t a s e s were p r e p a r e d a c c o r d i n g to Nilsson et al. ( 1 9 7 4 ) . T h e a m o u n t o f C3 b o u n d to the stable cell i n t e r m e d i a t e m a d e u p o f EA a n d the c o m p l e m e n t c o m p o n e n t s C4 a n d C2 ( i o d i n e - o x i d i z e d ) , f o r m i n g the classical C3 c o n v e r t a s e (EAC4°×Y2) was c o m p a r e d to the a m o u n t passively a d s o r b e d o n t o EA. V e r o n a l b u f f e r e d saline, 0.11 ml, c o n t a i n i n g 26 ng 125I-C3, 0.1% gelatin a n d 10 m M E D T A ( G V B - E D T A ) was i n c u b a t e d with 0.5 × 108 cells for 15 m i n at 3 7 ° C. A f t e r washing t h r e e times, the r a d i o a c t i v i t y was m e a s u r e d in a y-counter.
52
K.-E. STORM ET AL.
Hemolytic titration of C3 was performed using a modified procedure previously described by Mandle et al. ( 1977 ). All dilutions were made in GVBEDTA and the assays were carried out in microtiterplates (Immunoplates type I, NUNC, Roskilde, Denmark). Test sample ( 5 0 / d ) , at varying dilutions, and 50/A of potassium thiocyanate-treated guinea pig serum (GPKSCN ) diluted 1/32 and containing 0.1/tg m l - ' of purified human C5, were incubated at 37°C with 50 #1 of cell suspension ( 1 × 10 - 7 EAC4°XY2) for 60 min. Ice-cold veronal buffered saline ( 100 Ftl ), containing 0.05% Tween 20 and 0.02% antifoam (Pharmacia, Uppsala), was added. After 5 min of centrifugation at 225 × g, 100/~1 of the supernatants were analysed spectrophotometrically at 414 nm. For linearization of dose-response curves the z-transformation of the absorbance values was used; z = - I n (l-y) and y represents observed/total hemolysis (Borsos et al., 1964). Since the linear relation between concentration and hemolysis is distorted in crude preparations at low dilutions (Barta and Hubbert, 1978; swine serum less than 1/ 100), activity comparisons were made at dilutions within the linear dose-response area.
Proteolytic cleavage of C3 The fragmentation pattern for C3 was studied in specimens of whole serum incubated for 7 days at 37°C in the presence of 0.1% NaN3 (aged serum) (Bokisch et al., 1975 ), or in reaction mixtures containing trypsin and purified C3 (Bokisch et al., 1969).
Immunizations and immunochemical techniques New Zealand White rabbits were immunized according to Harboe and Ingild ( 1983 ) for human and porcine C3, and according to Nilsson et al. (1974) for human C5. These antibodies were used unabsorbed and as unseparated sera. Anti-human C4 and anti-swine serum proteins were obtained commercially (Behringwerke, Marburg, Germany and DAKOPATTS, Glostrup, Denmark). Double immunodiffusion, crossed immunoelectrophoresis and fused rocket electrophoresis were performed according to Wallenborg and Andersson (1978 ). Addition of PEG 6000 in the gels enhanced precipitability in reactions between antihuman C3, C4 and C5 antisera and their corresponding swine homologues (Kj~erwig Broe and Ingild, 1983 ).
Polyacrylamide gel electrophoresis Reduced and sodium dodecyl sulphate ( SDS )-denatured samples were run in 10% polyacrylamide gels (Laemmli and Favre, 1973 ).
PURIFICATION AND CHARACTERIZATION OF PORCINE C3
53
Amino acid analysis The amino acid analysis used a D u r r u m D-500 automatic amino acid analyser with a modified Moore-Stein system (Spackman et al., 1958 ). Half-cystein was analysed according to Moore (1963). For tryptophan analysis a nucleosil C 18 column was used. Published data from other species were compared against the amino acid composition of swine C3 using a linear regression model. The obtained r 2 values were used as percentage expressions of their relative homology. Data without a tryptophan estimate were completed by 10 tryptophan residues per 1000 amino acids and recalculated before comparison.
Carbohydrate analysis The carbohydrate analysis was performed according to Sawardeker et al. (1965), as described in a modification by Eggertsen et al. (1985). Neutral sugars including galactose- and glucoseamine but not ketoses (fructose, KDO etc. ) and acidic sugars (e.g. sialic acid) were detected. Mobilities of desialylated and untreated C3 were compared by high voltage agarose electrophoresis, kindly performed by Bo Ghane, SLU, Uppsala. RESULTS
Separation The DEAE-Sephacel elution profile and the distribution of C3, C4 and C5 (Figs. 1 and 2), illustrate the effectiveness of this separation step. Large molecular weight impurities were separated from C3 by Sepharose CL-6B chromatography (Fig. 3). A final step of purification by hydroxylapatite chromatography removed trace amounts of remaining C4, C5 and immunoglobulins (Fig. 4). C3 appeared in the eluate as a single homogeneous protein peak that corresponded completely to the distribution of C3 as measured functionally and antigenically against anti-whole swine serum, antiswine C3 or anti-human C3. SDS-PAGE analysis of the isolated C3 under reducing conditions yielded two protein bands presumed to represent the C3o~ and C3fl chains (compare with results below), without any contaminants. The recovery and specific activities of C3 in the intermediate and final steps of the purification are recorded in Table 1. On basis of an assumed extinction coefficient of 10.0 (280 nm, 1%, 1 cm), the final C3 pool contained 174 mg or 1.3% of the total protein, which corresponded to a 56% recovery of C3 in the starting plasma pool. Analysis by double immunodiffusion performed over wide ranges of antigen to antibody ratios (not shown), yielded a single precipitation band be-
54
K.-E. STORM ET AL.
tween purified swine C3 and rabbit anti-whole swine serum. Sera from two rabbits hyperimmunized against the purified swine C3, yielded single precipitin bands with fresh EDTA pig plasma. Crossed immunoelectrophoresis of fresh and aged serum at high antiserum concentrations failed to reveal precipitates inconsistent with the fragmentation of C3 (Fig. 5 ).
Comparativefunctional and structural studies of human and porcine C3 A partial antigenic identity between purified swine and human C3 was demonstrated by double immunodiffusion, both against rabbit anti-human C3 and rabbit anti-swine C3 antiserum (Fig. 6). SDS-PAGE of swine and h u m a n C3 under reducing conditions revealed a subunit composition with one o~- and one #-chain for both proteins (Fig. 7 ). The size of the a-chains is almost identical for the two proteins (approximately 110 kDa), while the #chain of the porcine protein was approximately 10 kDa lighter (approximately 60 kDa) than the h u m a n counterpart. The amino acid composition (Table 2) corresponds well to those previ-
Fresh s e r u m
Aged serum Fig. 5. Crossed i m m u n o e l e c t r o p h o r e s i s of fresh frozen a n d aged pig sera. The second dimensions are run in gelscontainingthe antisera raised againstthe purified swine C3, for aged serum
8.5/11cm-2 and for fresh serum 3.4/A cm-2.
PURIFICATION AND CHARACTERIZATION OF PORCINE C3
55
Hu C3 I / 4
P C3 114
P C3 I / 2
l.~u C3 I I I P C3 I / I -anti Hu-C31/1
Hu C3 I / 2 anti P-C31/1"
Fig. 6. Double immunodiffusion analysis of porcine and human C3.
I Retl 1 I 2 !
3
I 4 I s I 6 I 7 I 8 I
9
11o1111121Ref.I
Fig. 7. SDS-PAGE-analysis of reduced samples. Samples: purified human and porcine C3 with different degrees of degradation by trypsin treatment and reference proteins. Track 1, undigested human C3; Track 7, undigested porcine C3; Tracks 2-6, increasing degrees of trypsin digestion of human C3 by varying incubation times, 1, 5, 20, 60 and 180 rain respectively; Tracks 8-12, the corresponding treatment of porcine C3. ously published for h u m a n C3 ( D e Bruijn and Fey, 1985; 98.33%) and swine C3 (Pgtques, 1980; 94.79%). In agreement with data obtained for h u m a n C3 ( T o m a n a et al., 1985 ), swine C3 exhibited a moderate carbohydrate content (approximately 2%) (Table 3 ), with no indication of the presence of sialic acid as indicated by an unchanged electrophoretic mobility following desialylation. In studies of the hemolytic and binding functions, which used EAC4°Xy2 cells, the specific activities of porcine C3 were one-tenth and one-third that of h u m a n C3, respectively (Table 4).
Electrophoretic studies of C3 degradation products Analysis by crossed immunoelectrophoresis was performed on trypsin-digested purified C3 of the two species against anti-swine C3 or anti-human C3
56
K.-E. STORMET AL
TABLE 2 Amino acid composition. Residues per 1000 amino acids Amino acid
Residue
Asx Thr Ser Glx Pro Gly Ala Cys Val Met Ile Leu Tyr Phe His Lys Trp Arg Tot
101 57 63 127 50 63 56 18 95 14 50 88 33 35 20 72 11 46 999
TABLE 3 Carbohydrate analysis Carbohydrate
Residue per molecule
Galactose 4 Mannose 6 Glucose 10 N-acetylglucoseamin 1
TABLE 4 C3b-binding Amount (ng) Porcine C3 Human C3 Bound to EAC4°Xy2 Bound to EA
3.2 0.9
7,7 0,7
Difference
2.1
7,0
PURIFICATION AND CHARACTERIZATION OF PORCINE C3
57
antisera. While conversion into the electrophoretically faster C3b/C3c was readily demonstrable for human C3, a corresponding change was not seen with swine C3 (not shown). On the other hand, the fragmentation patterns as revealed by SDS-PAGE were analogous for porcine and human C3 with degradation of the c~-chains but with no effect on the fl-chains (Fig. 7 ). The final porcine C3 a-fragments of 29 and 25 kDa corresponded to the portions of the o~-chain present in human C3c and the stable porcine 33 kDa band to the fragment present in human C3d. Analysis of the acrylamide gels by immunoblotting revealed cross-reactions between the two former fragments, and anti-human C3c and the latter and anti-human C3d antiserum, respectively (not shown ). Crossed immunoelectrophoresis of trypsin-treated swine C3 against antiswine C3 antiserum, yielded a faint precipitate of C3d with similar mobility, and was almost covered by the main C3c peak (not shown). The analogous analysis of aged pig serum containing completely degraded C3 yielded a C3dg fragment of clearly more anodal electrophoretic mobility than C3c (Fig. 5 ). DISCUSSION
The isolation of highly purified and functionally intact swine C3 was accomplished by a modification of the method originally developed by Hammer et al. ( 1981 ) for the preparation of human complement proteins. The initial preparation steps involving PEG precipitation, plasminogen depletion and DEAE Sephacel and Sepharose CL-6B chromatography were performed according to the original procedure, while hydroxylapatite was used instead of affinity chromatography for the final purification step. Helpful in the handling of the swine proteins and probably essential to the successful isolation of functionally intact porcine C3 at high yields, was the application of techniques to support protein solubility throughout the procedure. Protein precipitation at low ionic strength in the starting buffer for DEAE Sephacel chromatography was prevented by the combined fast buffer exchange and PEG removal accomplished by the Sephadex G-25 gel filtration. Efficient resolubilization of proteins following precipitation/concentration in 16% PEG was obtained by adding Tween 20 to the resolubilizing buffer. Immunoprecipitation against cross-reacting antisera in agarose gels with PEG-containing medium established the elution profile for C3, C4 and C5 in the different chromatographic procedures. The protein elution profile and the distribution of the individual proteins of the porcine complement system in the DEAE Sephacel eluate were in good agreement with those observed for the human system (Hammer et al., 1981 ). Measurements of the specific activity of C3 in the various intermediate fractions of the separation procedure (Table 1 ) revealed an initial rise above the C3 activity level of the original plasma pool, which was followed by a decrease in the activity of the two final steps of the purification. This rise may
58
K.-E. STORM ET AL.
well be related to the elimination of swine components, which compete/interfere with the heterologous components of the functional assay (Barta and Hubbert, 1978). The decrease of activity that occurred in the C3 final pool did not coincide with any demonstrable degradation of the protein and appears to be due to the breakage of the internal thiolester bond of C3 and the formation of iC3 by the interaction with surrounding water during handling (Pangburn and Miiller-Eberhard, 1980). The functional analysis of swine C3 was performed in a reaction mixture composed of human complement factors in an assay designed for the analysis of human C3. In the C3b binding studies, the efficiency of swine C3 was onethird that of the human analogue. A correction of this value for the observed loss (27%) of swine C3 function during preparation would bring its true efficiency to some 46% of human C3. The low specific hemolytic activity of swine C3 (one-tenth of that seen in humans) points to a more pronounced incompatibility between swine and human complement in the reaction steps following C3 binding, and a reduction of the swine C3 function in this step to some 30% in the heterologous test system. Examples of inefficient reactions in the presence of heterologous components have been published for several species combinations (Barta and Hubbert, 1978; Linscott, 1986). Evidence of a high degree of homology between human and porcine C3 was provided by their partial antigenic identity (Fig. 6 ) and their similar amino acid composition (more than 98%) and carbohydrate content (approximately 2%). Findings supporting this are also provided by the studies of the polypeptide chain subunit arrangement and the proteolytic degradation patterns of the a-chains of the two species. Also, some potentially important differences between porcine and human C3 were noted. First, the swine C3 fl-chain at 60 kDa is 10 kDa lighter than its human homologue. A similar unexplained fl-chain difference is apparent in mice and guinea pigs (Nonaka et al., 1980; Thomas and Tack, 1983 ), there the N-terminal analysis revealed a conserved sequence between these species and the human. The difference is too large to be explained by the lack of carbohydrates in the mouse fl-chain. In addition, the nucleotide-derived sequences of the complete chain confirms the lack of carbohydrate, but reveals also that the mouse fl-chain is only three amino acids shorter than the human chain (642 and 645 amino acids) (Lundwall et al., 1984; De Bruijn and Fey, 1985). Second, swine C3b/iC3b/C3c generated by trypsin digestion of purified C3 or by storage of serum, have electrophoretic mobilities very close to that of intact C3. Consequently, and in contrast to the human condition (Bokisch et al., 1969), swine C3 fragments cannot easily be detected by crossed immunoelectrophoresis and used as indicators of complement activation. This may explain the failure to demonstrate C3 split products in synovial fluid of arthritic swine by immunoelectrophoresis (Timoney, 1976). On the other
PURIFICATION AND CHARACTERIZATION OF PORCINE C3
59
hand, the electrophoretic properties of C3dg in aged serum and oftryptic C3d are similar to the human counterparts (Lachmann et al., 1982), and in contrast to C3b/iC3b/C3c, the C3d fragment seems equally well-suited as a marker for the activation of porcine as well as human complement (Nilsson et al., 1990). This investigation has improved the separation techniques for swine C3 and indicated possible preparative procedures for the isolation of C4 and C5 of the same species. Knowledge has been extended regarding the characteristics of some activation products of swine C3 and a previously unrecognized difference in molecular weight between human and swine C3 fl-chain has been pointed out. Also, an antigenic cross-reaction between human and swine C3 is shown that could not be demonstrated by earlier preparations or reagents, and the herein reported amino acid composition correlates more strongly with human C3 than with earlier swine C3 determinations. ACKNOWLEDGEMENTS
We are deeply grateful to the late Professor Bo Ghane, Department of Animal Breeding and Genetics, SLU, Uppsala, who generously shared his great knowledge with us, and kindly performed the electrophoretic study of neuraminidase-treated C3. We are also thankful to Per-Erik Jansson, Department of Organic Chemistry, Arrhenius Laboratory, University of Stockholm, for kindly performing the carbohydrate analysis, Bo Nilsson for critical reading of the manuscript, Margita Nilsson and Susanne Heller for skillful technical assistance. This work was supported by grants from the Swedish Medical Research Council (no. 5674), the Swedish National Defence Research Institute, and from the Swedish National Board of Technical Development.
REFERENCES Barta, O. and Hubbert, N.L., 1978. Testing of hemolytic complement components in domestic animals. Am. J. Vet. Res., 39: 1303-1308. Bokisch, V.A., Dierich, M.P. and Mtiller-Eberhard, H.J., 1975. Third component of complement (C3): structural properties in relation to functions. Proc. Natl. Acad. Sci. USA, 72: 1989-1993. Bokisch, V.A., Miiller-Eberhard, H.J. and Cochrane, C.G., 1969. Isolation of a fragment (C3a) of the third component of human complement containing anaphylatoxin and chemotactic activity and description of an anaphylatoxin inactivator of human serum. J. Exp. Med., 129: 1109-1130. Borg, T., Alvfors, A., Gerdin, B. and Modig, J., 1985. A porcine model of early adult respiratory distress syndrome induced by endotoxemia. Acta Anaesthesiol. Scand., 29:814-830. Borsos, T., Dourmaskin, R.R. and Humphrey, J.H., 1964. Lesion in erythrocyte membranes caused by immune hemolysis. Nature, 202:251-252.
60
K.-E. S T O R M ET AL.
Cooper, N.R., 1988. Laboratory investigations of complement proteins and complement receptors. In: M.D. Kazatchkine and T. McMonagle (Editors), Complement and Immunological Disease. 2. Bailli6re Tindall, London, pp. 263-293. De Bruijn, M.H.L. and Fey, G.H., 1985. Human complement component C3: cDNA coding sequence and derived primary structure. Proc. Natl. Acad. Sci. USA, 82:708-712. Eggertsen, G., Hellman, U., Lundwall, ]~., Folkersen, J. and Sj6quist, J., 1985. Characterization of tryptic fragments of human complement factor C3. Mol. Immunol., 22:833-841. Hammer, C.H., Wirtz, G.H., Renfer, L., Gresham, H.D. and Tack, B.F., 1981. Large scale isolation of functionally active components of the human complement system. J. Biol. Chem., 256: (8): 3995-4006. Harboe, N.M.G. and Ingild, A., 1983. Immunization, isolation of immunoglobulins and antibody titre determination. In: N.H. Axelsen (Editor), Handbook of Immunoprecipitationin-Gel Techniques. Scandinavian Journal of Immunology. Supplement 10, Vol. 17. Blackwell, Oxford, UK, pp. 345-351. Kjaerwig Broe, M. and Ingild, A., 1983. Amplification ofimmunoprecipitates in agarose gels by horseradish peroxidase-labeled antibody. In: N.H. Axelsen (Editor), Handbook of Immunoprecipitation-in-Gel Techniques. Scandinavian Journal of Immunology. Supplement 10, Vol. 17. Blackwell, pp. 255-258. Lachmann, P.J., Pangburn, M.K. and Oldroyd, R.G., 1982. Breakdown of C3 after complement activation: identification of a new fragment, C3g, using monoclonal antibodies. J. Exp. Med., 156: 205-216. Laemmli, U.K. and Favre, M., 1973. Maturation of the head of bacteriophage T4. J. Mol. Biol., 80: 575-599. Lambris, J.D., 1988. The multifunctional role of C3, the third component of complement. Immunol. Today, 9( 12): 387-393. Linscott, W.D., 1986. Biochemistry and biology of the complement system in domestic animals. Prog. Vet. Microbiol. Immunol., 2: 54-77. Lundwall, ]~., Wetsel, R.A., Domdey, H., Tack, B.F. and Fey, G.H., 1984. Structure of murine complement component C3. I. Nucleotide sequence of cloned complementary and genomic DNA coding for the beta chain. J. Biol. Chem., 259:13851-13856. Mandle, R.J., McConnel-Mapes, J.A. and Nilsson, U.R., 1977. Inhibition of complement by mouse serum: selective inactivation of the fourth component of human complement. J. Immunol., 119(1): 180-186. Moore, S., 1963. On the determination of cystine as cysteic acid. J. Biol. Chem., 238( 1 ): 235237. Miiller-Eberhard, H.J., 1988. Molecular organization and function of the complement system. Annu. Rev. Biochem., 57: 321-347. Nilsson, B. and Nilsson, U.R., 1985. An assessment of the extent of antigenic analogy between physiologically bound C3 and C3 denatured by sodium dodecyl sulphate. Scand. J. Immunol., 22(6): 703-710. Nilsson, U.R., Miller, M.E. and Wyman, S., 1974. A functional abnormality of the fifth component of complement (C5) from human serum of individuals with a familial opsonic defect. J. Immunol., 112(3): 1164-1176. Nilsson, L., Storm, K.-E., Thelin, S., Bagge, L., Hultman, J., Thorelius, J. and Nilsson, U., 1990. Heparin-coated equipment reduces complement activation during cardiopulmonary bypass in the pig. Artif. Organs, 14( 1 ): 46-48. Nilsson, U.R., Tomar, R.H. and Taylor, F.B., 1972. Additional studies on human C5: development of a modified purification method and characterization of the purified product by polyacrylamide gel electrophoresis. Immunochemistry, 9: 709-723. Nonaka, M., Natsuume-Sakai, S., Kaidoh, T., Nonaka, M. and Takahashi, M., 1980. Characterization of allotypic determinants of murine complement C3. J. Immunol., 125: 2025-2030.
PURIFICATION AND CHARACTERIZATION OF PORCINE C3
61
Pangburn, M.K. and Mtiller-Eberhard, H.J., 1980. Relation of a putative thioesterbond in C3b to activation of the alternative pathway and the binding of C3b to biological targets of complement. J. Exp. Med., 152:1102-I 114. Pfiques, E.P., 1980. Purification and partial characterization of the third component of the complement system from porcine serum (C3) and of a crystallizable degradation product of the fourth component of the complement system from human serum (C4). Hoppe-Seyler's Z. Physiol. Chem., 361: 445-456. Sawardeker, J.S., Sloneker, J.H. and Jeanes, A., 1965. Quantitative determination of monosaccharides as their alditol acetates by gas liquid chromatography. Anal. Chem., 37: (12) 16021604. Spackman, D.H., Stein, W.H. and Moore, S., 1958. Automatic recording apparatus for use in the chromatography of amino acids. Anal. Chem., 30 (7): 1190-1206. Sundsmo, J.S. and Fair, D.S., 1985. Relationships among the complement, kinin, coagulation, and fibrinolytic systems. In: H.J. Miiller-Eberhard and P.A. Miescher (Editors), Complement. Springer, Berlin, pp. 379-406. Thomas, M.L. and Tack, B.F., 1983. Identification and alignment of a thiol ester site in the third component of guinea pig complement. Biochemistry, 22 (4): 942-947. Timoney, J.F., 1976. Erysipelas arthritis in swine: concentrations of complement and third component of complement in synovia. Am. J. Vet. Res., 37( 1 ): 5-8. Tomana, M., Niemann, M., Garner, C. and Volanakis, J.E., 1985. Carbohydrate composition of the second, third and fifth components and factors B and D of human complement. Mol. lmmunol., 22:107-111. Wallenborg, B. and Andersson, U.-B., 1978. Immunoelectrophoretic Techniques with the LKB 2117 Multiphor. Application Note 249, LKB-Produkter, Bromma, Sweden, 12 pp. Zidkov~, J., V~tvi~ka, V., Rossman, P., Dlaba~, V. and Strfinsky, J., 1984. Characterization of the third component of pig complement and its utilization in a C3b receptor study. Experientia, 40: 975-977.
47
Purification and characterization of porcine C3. Studies of the biologically active protein and its split products K a r l - E r i k S t o r m a, G 6 s t a A r t u r s o n b a n d U l f R . N i l s s o n a aDepartment of Clinical Immunology and Transfusion Medicine, UniversityHospital S- 751 85 Uppsala, Sweden bBurn Center, UniversityHospital S-751 85 Uppsala, Sweden (Accepted 28 November 1991 )
ABSTRACT Storm, K.-E., Arturson, G. and Nilsson, U.R., 1992. Purification and characterization of porcine C3. Studies of the biologically active protein and its split products. Vet. Immunol. Immunopathol., 34: 47-61. Separation techniques for obtaining pure and biologically active swine C3 have been improved in this study. Using these procedures and through the further characterization of porcine C3, the possibilities for developing more specific techniques for the analysis of the complement system in swine have been improved. Plasma was initially treated with protease inhibitors, polyethylene glycol (PEG)fractionation, plasminogen-depletionand a rapid chromatographic desalting step. The essential fractionation was carried out by DEAE-Sephacel chromatography. Contaminants were removed by sizeexclusion ( Sepharose CL-6B )- and hydroxylapatite-chromatography. The final recovery reached 56% with 73% retaining specific hemolytic activity. The amino acid composition (98.33%), the functional compatibility and the secondary structure of fragments and intact protein indicate a high degree of homology with human C3. In contrast with the findings of earlier studies was the considerable immunologic cross-reactivity observed with human C3, and the size difference between the human and the swine C3-fl subunit, which was found to be 10 kDa lighter than the human analogue. The finding that the swine C3b/iC3b/C3c fragments do not separate from C3 by agarose electrophoresis, unlike the human analogues, demonstrated that this commonly used simple parameter for the detection of complement activation cannot be used in the porcine model. ABBREVIATION EA, sheep erythrocytes sensitized with rabbit hemolysin; EAC4°Xy2, the stable cell intermediate made up of EA and the complement components C4 and C2 (iodine-oxidized), forming the classical third component of complement (C3) convertase; PEG, polyethylene glycol; SDS, sodium dodecyl sulphate; SDS-PAGE, sodium dodecyl sulphate-polyacrylamide gel electrophoresis.
Correspondence to: Karl-Erik Storm, Department of Clinical Immunology and Transfusion Medicine, University Hospital, S-751 85 Uppsala, Sweden.
© 1992 Elsevier Science Publishers B.V. All rights reserved 0165-2427/92/$05.00
48
K.-E. STORM ET AL.
INTRODUCTION
Complement is an important effector system which is involved in the host defense and in numerous vital reactions of the inflammatory process, e.g. vascular permeability, opsonization of foreign cells and proteins and regulation of the immune response. It comprises more than 30 plasma and membrane proteins, which are organized in two activation pathways, one predominantly dependent, the other independent of antibodies, and which both trigger the various biologic effects of the common terminal reaction sequence (MiillerEberhard, 1988). C3 is the most abundant complement protein and participates in both activation pathways. Specific interactions with cell surface molecules, many serum proteins and proteins of foreign origin indicate the presence on C3 of at least ten specific binding sites. The fragmentation during activation modulates the expression of these binding domains, which are further tuneable both by the action of co-factors and by the covalent linkage of the molecule by its reactive internal thiolester (Lambris, 1988 ). Several new techniques have been designed for the analysis of the human complement system, and by the assessment of its physicochemical alterations, activation can be quantified precisely (Cooper, 1988 ). These new techniques therefore increase the possibilities of evaluating the relative importance of complement compared t o other effector systems such as the coagulation-fibrinolytic andkinin-gener~iting systems (Sundsmo and Fair, 1985). The techniques for analysis of the porcine complement system are far less developed as reflected in recent studies of inflammation which limit the complement analysis to the hemolytic test of the entire complement sequence. New analytical procedures have to be developed to make this model available for complement analysis at a level comparable to that possible for the human. An essential step is to make isolated components and fragments of the swine complement system available for specific analysis. Although several porcine complement proteins have been isolated, little effort has been made to study their reaction in the homologous system (Linscott, 1986). In view of the lack of information on the specific activity of purified swine C3 in earlier publications (P~tques, 1980; Zidkov~i et al., 1984), it seemed justified to apply more efficient techniques (Nilsson et al., 1972; Hammer et al., 1981 ) to the isolation, identification and partial characterization of biologically active porcine C3 in the present report. MATERIALS AND METHODS
Approximately 15-week-old pigs were exsanguinated under general anaesthesia with Ketamin as described (Borg et al., 1985 ). The blood was collected
PURIFICATION AND CHARACTERIZATIONOF PORCINE C3
49
in plastic bags containing anti-coagulant (CPD-Sagman, Terumo, Pharmacia, Uppsala). After separation of cells by centrifugation the plasma was kept frozen at - 70 ° C.
C3 separation procedure The sequential steps of the separation procedure can be followed in the left column of Table 1. Swine plasma was treated with inhibitors and polyethylene glycol (PEG) in accordance with the initial anti-proteolytic procedure described by Hammer et al. ( 1981 ). The soluble fraction was applied to a Lysine-Sepharose R (Pharmacia, Uppsala) column ( 5 × 5.5 cm, 195 ml h - 1). The bulk of eluted protein was pooled and transferred by gel filtration on Sephadex R G-25 (Pharmacia, Uppsala) ( 10 × 28.8 cm, 2647 ml h - ~) to buffer A and applied to the DEAE-Sephacel R (Pharmacia, Uppsala) column (Fig. 1 ). Eluted fractions containing C3 were identified by rocket electrophoresis and by hemolytic C3 screening (Figs. 1 and 2). Appropriate fractions were pooled and concentrated by PEG-precipitation according to Hammer et al. (1981). The precipitate was dissolved in approximately 20 ml equilibration buffer containing 0.05% Tween-20, and separated by Sepharose R CL-6B (Pharmacia, Uppsala) chromatography (Fig. 3). The eluate was analysed hemolytically and by fused rocket electophoresis against anti-swine C3 and anti-swine serum proteins respectively. C3 containing fractions were pooled and concentrated. The resulting precipitate of the Sepharose CL-6B pool was dissolved in approximately 20 ml buffer A (Fig. 4 ) containing 0.05% Tween 20, and applied to a column of hydroxylapatite (DNA-grade, Bio-Rad, Richmond, CA) under conditions modified (Fig. 4) from Nilsson et al. (1972). After similar TABLE 1 Summary of yields Fraction
Volume Total protein pool (ml) Conc. Recovery (mgm1-1) (%)
Plasma 360 PEG-supernatant 400 Lysine-Sepharose 460 Sephadex G-25 1010 DEAE-Sephacel 378 Sepharose CL-6B 71.5 Hydroxylapatite 119
37 34 29 12 1.4 3.4 1.5
100 101 99 91 3.9 1.8 1.3
Immunochemical activity
C3 activity
Conc. (mg ml -I )
Recovery Conc. Specific Recovery (%) ( z m l -~) (z [C3] -~) (%)
0.99 0.84 0.76 0.33 0.79 3.3 1.7
100 94 98 92 83 66 56
296 241 169 109 319 1548 365
298 287 223 333 405 471 218
100 90 73 102 113 104 41
50
K,-E. STORM ET AL.
% z
A 28°
mS
so ~
13 wash
i i ........
1
0 1 gradient
2
i
,o
3
v o l u m e (I)
Fig. 1. Anion exchange chromatography of Sephadex G-25 pool. Conditions: column, DEAESephacel (5 × 51.4 cm); sample size 1010 ml; buffer A, 3.2 mM K/Na/H/PO4, 6.4 mM EDTA, 6.4 mM benzamidine-HCl, 31.8 mM EACA, adjusted to 3.0 mS + 4 ° C with NaC1, pH 7.4; buffer B, buffer A adjusted to 18.0 mS+4°C with NaC1; gradient, 620 ml buffer A, 3000 ml 0-100% B, 1000 ml buffer B; flow rate 120 ml h- ~;fraction volume, 10 ml; pooled material, fraction 188 through 225; detection, absorbance 280 nm ( ), conductivity (mS) at + 4°C ( ..... ), relative C3-hemolytic activity (- - -), 100%=maximum z.
anti-C5
anti-C3
............
anti-C4
000000000000000000000000 0000000000000000000000000 I
1000
I
I
2000
I
I
3000
I
v o l u m e (ml)
Fig. 2. Fused rocket immunoelectrophoresis of eluted fractions from DEAE-Sephacel. C3 is detected with specific anti-swine C3. Cross-reacting anti-human antisera are used for the C4 and C5 detection.
screening as for the eluate o f the p r e c e d i n g c o l u m n , the p e a k was p o o l e d a n d concentrated. H u m a n C3 was p r e p a r e d using the s a m e p r o c e d u r e s with the e x c e p t i o n t h a t the a d d i t i o n o f T w e e n 20 in the s o l u b i l i z a t i o n o f P E G - p r e c i p i t a t e s was omitted.
PURIFICATION AND CHARACTERIZATION OF PORCINE C3
0
260
460
660
51
volume (ml)
Fig. 3. Gel filtration of the DEAE-Sephacel pool. Conditions: column, Sepharose CL-6B (2.6 × 95 cm); sample size 21.5 ml; buffer, 100 mM K/Na/H/PO4, 150 mM NaC1, 5 mM EDTA, 50 mM EACA, pH 7.4; flow rate 20 ml h- ~h; fraction volume, 5 ml; pooled material, fraction 61 through 75; detection, absorbance 280 nm ( ), C3-hemolytic activity in percent z-units of highest value (- - -). -,o A 280 100 3
50
mS f 20
21
wash
i ~ i ....... ....
0 gradient
500
10
1000 volume (roll
Fig. 4. Hydroxylapatite chromatography of Sepharose CL-6B pool. Conditions: DNA-grade hydroxylapatite (5 × 10 cm); sample size 18.8 ml; buffer A, 20 mM K/Na/H/PO4, 100 mM EACA, adjusted with buffer C to 7.5 mS+4°C, pH 7.9; buffer B, 20 mM K/Na/H/PO4, 100 mM EACA, adjusted with buffer C to 20.0 mS + 4 ° C, pH 7.9; buffer C, 650 mM K/Na/H/PO4, 100 mM EACA, pH 7.9; gradient, 300 ml buffer A, 1000 ml 0-100% B; flow rate 30 ml h-~; fraction volume 5 ml; pooled material, fraction 81 through 105; detection, absorbance 280 nm ( ), conductivity (mS) at + 4 °C ( ..... ), C3-hemolytic activity in percent z-units of highest value (- - -).
C3b-binding and hemolytic techniques C o n v e r t a s e - d e p e n d e n t C 3 b - b i n d i n g was assessed with 125I-labelled C3. Labelling o f C3 with 125I ( A m e r s h a m Sweden, S o l n a ) to a final activity o f 1.5 × 106 c p m / t g - 1 was p e r f o r m e d a c c o r d i n g to N i l s s o n a n d Nilsson ( 1985 ). EA a n d c e l l - b o u n d c o n v e r t a s e s were p r e p a r e d a c c o r d i n g to Nilsson et al. ( 1 9 7 4 ) . T h e a m o u n t o f C3 b o u n d to the stable cell i n t e r m e d i a t e m a d e u p o f EA a n d the c o m p l e m e n t c o m p o n e n t s C4 a n d C2 ( i o d i n e - o x i d i z e d ) , f o r m i n g the classical C3 c o n v e r t a s e (EAC4°×Y2) was c o m p a r e d to the a m o u n t passively a d s o r b e d o n t o EA. V e r o n a l b u f f e r e d saline, 0.11 ml, c o n t a i n i n g 26 ng 125I-C3, 0.1% gelatin a n d 10 m M E D T A ( G V B - E D T A ) was i n c u b a t e d with 0.5 × 108 cells for 15 m i n at 3 7 ° C. A f t e r washing t h r e e times, the r a d i o a c t i v i t y was m e a s u r e d in a y-counter.
52
K.-E. STORM ET AL.
Hemolytic titration of C3 was performed using a modified procedure previously described by Mandle et al. ( 1977 ). All dilutions were made in GVBEDTA and the assays were carried out in microtiterplates (Immunoplates type I, NUNC, Roskilde, Denmark). Test sample ( 5 0 / d ) , at varying dilutions, and 50/A of potassium thiocyanate-treated guinea pig serum (GPKSCN ) diluted 1/32 and containing 0.1/tg m l - ' of purified human C5, were incubated at 37°C with 50 #1 of cell suspension ( 1 × 10 - 7 EAC4°XY2) for 60 min. Ice-cold veronal buffered saline ( 100 Ftl ), containing 0.05% Tween 20 and 0.02% antifoam (Pharmacia, Uppsala), was added. After 5 min of centrifugation at 225 × g, 100/~1 of the supernatants were analysed spectrophotometrically at 414 nm. For linearization of dose-response curves the z-transformation of the absorbance values was used; z = - I n (l-y) and y represents observed/total hemolysis (Borsos et al., 1964). Since the linear relation between concentration and hemolysis is distorted in crude preparations at low dilutions (Barta and Hubbert, 1978; swine serum less than 1/ 100), activity comparisons were made at dilutions within the linear dose-response area.
Proteolytic cleavage of C3 The fragmentation pattern for C3 was studied in specimens of whole serum incubated for 7 days at 37°C in the presence of 0.1% NaN3 (aged serum) (Bokisch et al., 1975 ), or in reaction mixtures containing trypsin and purified C3 (Bokisch et al., 1969).
Immunizations and immunochemical techniques New Zealand White rabbits were immunized according to Harboe and Ingild ( 1983 ) for human and porcine C3, and according to Nilsson et al. (1974) for human C5. These antibodies were used unabsorbed and as unseparated sera. Anti-human C4 and anti-swine serum proteins were obtained commercially (Behringwerke, Marburg, Germany and DAKOPATTS, Glostrup, Denmark). Double immunodiffusion, crossed immunoelectrophoresis and fused rocket electrophoresis were performed according to Wallenborg and Andersson (1978 ). Addition of PEG 6000 in the gels enhanced precipitability in reactions between antihuman C3, C4 and C5 antisera and their corresponding swine homologues (Kj~erwig Broe and Ingild, 1983 ).
Polyacrylamide gel electrophoresis Reduced and sodium dodecyl sulphate ( SDS )-denatured samples were run in 10% polyacrylamide gels (Laemmli and Favre, 1973 ).
PURIFICATION AND CHARACTERIZATION OF PORCINE C3
53
Amino acid analysis The amino acid analysis used a D u r r u m D-500 automatic amino acid analyser with a modified Moore-Stein system (Spackman et al., 1958 ). Half-cystein was analysed according to Moore (1963). For tryptophan analysis a nucleosil C 18 column was used. Published data from other species were compared against the amino acid composition of swine C3 using a linear regression model. The obtained r 2 values were used as percentage expressions of their relative homology. Data without a tryptophan estimate were completed by 10 tryptophan residues per 1000 amino acids and recalculated before comparison.
Carbohydrate analysis The carbohydrate analysis was performed according to Sawardeker et al. (1965), as described in a modification by Eggertsen et al. (1985). Neutral sugars including galactose- and glucoseamine but not ketoses (fructose, KDO etc. ) and acidic sugars (e.g. sialic acid) were detected. Mobilities of desialylated and untreated C3 were compared by high voltage agarose electrophoresis, kindly performed by Bo Ghane, SLU, Uppsala. RESULTS
Separation The DEAE-Sephacel elution profile and the distribution of C3, C4 and C5 (Figs. 1 and 2), illustrate the effectiveness of this separation step. Large molecular weight impurities were separated from C3 by Sepharose CL-6B chromatography (Fig. 3). A final step of purification by hydroxylapatite chromatography removed trace amounts of remaining C4, C5 and immunoglobulins (Fig. 4). C3 appeared in the eluate as a single homogeneous protein peak that corresponded completely to the distribution of C3 as measured functionally and antigenically against anti-whole swine serum, antiswine C3 or anti-human C3. SDS-PAGE analysis of the isolated C3 under reducing conditions yielded two protein bands presumed to represent the C3o~ and C3fl chains (compare with results below), without any contaminants. The recovery and specific activities of C3 in the intermediate and final steps of the purification are recorded in Table 1. On basis of an assumed extinction coefficient of 10.0 (280 nm, 1%, 1 cm), the final C3 pool contained 174 mg or 1.3% of the total protein, which corresponded to a 56% recovery of C3 in the starting plasma pool. Analysis by double immunodiffusion performed over wide ranges of antigen to antibody ratios (not shown), yielded a single precipitation band be-
54
K.-E. STORM ET AL.
tween purified swine C3 and rabbit anti-whole swine serum. Sera from two rabbits hyperimmunized against the purified swine C3, yielded single precipitin bands with fresh EDTA pig plasma. Crossed immunoelectrophoresis of fresh and aged serum at high antiserum concentrations failed to reveal precipitates inconsistent with the fragmentation of C3 (Fig. 5 ).
Comparativefunctional and structural studies of human and porcine C3 A partial antigenic identity between purified swine and human C3 was demonstrated by double immunodiffusion, both against rabbit anti-human C3 and rabbit anti-swine C3 antiserum (Fig. 6). SDS-PAGE of swine and h u m a n C3 under reducing conditions revealed a subunit composition with one o~- and one #-chain for both proteins (Fig. 7 ). The size of the a-chains is almost identical for the two proteins (approximately 110 kDa), while the #chain of the porcine protein was approximately 10 kDa lighter (approximately 60 kDa) than the h u m a n counterpart. The amino acid composition (Table 2) corresponds well to those previ-
Fresh s e r u m
Aged serum Fig. 5. Crossed i m m u n o e l e c t r o p h o r e s i s of fresh frozen a n d aged pig sera. The second dimensions are run in gelscontainingthe antisera raised againstthe purified swine C3, for aged serum
8.5/11cm-2 and for fresh serum 3.4/A cm-2.
PURIFICATION AND CHARACTERIZATION OF PORCINE C3
55
Hu C3 I / 4
P C3 114
P C3 I / 2
l.~u C3 I I I P C3 I / I -anti Hu-C31/1
Hu C3 I / 2 anti P-C31/1"
Fig. 6. Double immunodiffusion analysis of porcine and human C3.
I Retl 1 I 2 !
3
I 4 I s I 6 I 7 I 8 I
9
11o1111121Ref.I
Fig. 7. SDS-PAGE-analysis of reduced samples. Samples: purified human and porcine C3 with different degrees of degradation by trypsin treatment and reference proteins. Track 1, undigested human C3; Track 7, undigested porcine C3; Tracks 2-6, increasing degrees of trypsin digestion of human C3 by varying incubation times, 1, 5, 20, 60 and 180 rain respectively; Tracks 8-12, the corresponding treatment of porcine C3. ously published for h u m a n C3 ( D e Bruijn and Fey, 1985; 98.33%) and swine C3 (Pgtques, 1980; 94.79%). In agreement with data obtained for h u m a n C3 ( T o m a n a et al., 1985 ), swine C3 exhibited a moderate carbohydrate content (approximately 2%) (Table 3 ), with no indication of the presence of sialic acid as indicated by an unchanged electrophoretic mobility following desialylation. In studies of the hemolytic and binding functions, which used EAC4°Xy2 cells, the specific activities of porcine C3 were one-tenth and one-third that of h u m a n C3, respectively (Table 4).
Electrophoretic studies of C3 degradation products Analysis by crossed immunoelectrophoresis was performed on trypsin-digested purified C3 of the two species against anti-swine C3 or anti-human C3
56
K.-E. STORMET AL
TABLE 2 Amino acid composition. Residues per 1000 amino acids Amino acid
Residue
Asx Thr Ser Glx Pro Gly Ala Cys Val Met Ile Leu Tyr Phe His Lys Trp Arg Tot
101 57 63 127 50 63 56 18 95 14 50 88 33 35 20 72 11 46 999
TABLE 3 Carbohydrate analysis Carbohydrate
Residue per molecule
Galactose 4 Mannose 6 Glucose 10 N-acetylglucoseamin 1
TABLE 4 C3b-binding Amount (ng) Porcine C3 Human C3 Bound to EAC4°Xy2 Bound to EA
3.2 0.9
7,7 0,7
Difference
2.1
7,0
PURIFICATION AND CHARACTERIZATION OF PORCINE C3
57
antisera. While conversion into the electrophoretically faster C3b/C3c was readily demonstrable for human C3, a corresponding change was not seen with swine C3 (not shown). On the other hand, the fragmentation patterns as revealed by SDS-PAGE were analogous for porcine and human C3 with degradation of the c~-chains but with no effect on the fl-chains (Fig. 7 ). The final porcine C3 a-fragments of 29 and 25 kDa corresponded to the portions of the o~-chain present in human C3c and the stable porcine 33 kDa band to the fragment present in human C3d. Analysis of the acrylamide gels by immunoblotting revealed cross-reactions between the two former fragments, and anti-human C3c and the latter and anti-human C3d antiserum, respectively (not shown ). Crossed immunoelectrophoresis of trypsin-treated swine C3 against antiswine C3 antiserum, yielded a faint precipitate of C3d with similar mobility, and was almost covered by the main C3c peak (not shown). The analogous analysis of aged pig serum containing completely degraded C3 yielded a C3dg fragment of clearly more anodal electrophoretic mobility than C3c (Fig. 5 ). DISCUSSION
The isolation of highly purified and functionally intact swine C3 was accomplished by a modification of the method originally developed by Hammer et al. ( 1981 ) for the preparation of human complement proteins. The initial preparation steps involving PEG precipitation, plasminogen depletion and DEAE Sephacel and Sepharose CL-6B chromatography were performed according to the original procedure, while hydroxylapatite was used instead of affinity chromatography for the final purification step. Helpful in the handling of the swine proteins and probably essential to the successful isolation of functionally intact porcine C3 at high yields, was the application of techniques to support protein solubility throughout the procedure. Protein precipitation at low ionic strength in the starting buffer for DEAE Sephacel chromatography was prevented by the combined fast buffer exchange and PEG removal accomplished by the Sephadex G-25 gel filtration. Efficient resolubilization of proteins following precipitation/concentration in 16% PEG was obtained by adding Tween 20 to the resolubilizing buffer. Immunoprecipitation against cross-reacting antisera in agarose gels with PEG-containing medium established the elution profile for C3, C4 and C5 in the different chromatographic procedures. The protein elution profile and the distribution of the individual proteins of the porcine complement system in the DEAE Sephacel eluate were in good agreement with those observed for the human system (Hammer et al., 1981 ). Measurements of the specific activity of C3 in the various intermediate fractions of the separation procedure (Table 1 ) revealed an initial rise above the C3 activity level of the original plasma pool, which was followed by a decrease in the activity of the two final steps of the purification. This rise may
58
K.-E. STORM ET AL.
well be related to the elimination of swine components, which compete/interfere with the heterologous components of the functional assay (Barta and Hubbert, 1978). The decrease of activity that occurred in the C3 final pool did not coincide with any demonstrable degradation of the protein and appears to be due to the breakage of the internal thiolester bond of C3 and the formation of iC3 by the interaction with surrounding water during handling (Pangburn and Miiller-Eberhard, 1980). The functional analysis of swine C3 was performed in a reaction mixture composed of human complement factors in an assay designed for the analysis of human C3. In the C3b binding studies, the efficiency of swine C3 was onethird that of the human analogue. A correction of this value for the observed loss (27%) of swine C3 function during preparation would bring its true efficiency to some 46% of human C3. The low specific hemolytic activity of swine C3 (one-tenth of that seen in humans) points to a more pronounced incompatibility between swine and human complement in the reaction steps following C3 binding, and a reduction of the swine C3 function in this step to some 30% in the heterologous test system. Examples of inefficient reactions in the presence of heterologous components have been published for several species combinations (Barta and Hubbert, 1978; Linscott, 1986). Evidence of a high degree of homology between human and porcine C3 was provided by their partial antigenic identity (Fig. 6 ) and their similar amino acid composition (more than 98%) and carbohydrate content (approximately 2%). Findings supporting this are also provided by the studies of the polypeptide chain subunit arrangement and the proteolytic degradation patterns of the a-chains of the two species. Also, some potentially important differences between porcine and human C3 were noted. First, the swine C3 fl-chain at 60 kDa is 10 kDa lighter than its human homologue. A similar unexplained fl-chain difference is apparent in mice and guinea pigs (Nonaka et al., 1980; Thomas and Tack, 1983 ), there the N-terminal analysis revealed a conserved sequence between these species and the human. The difference is too large to be explained by the lack of carbohydrates in the mouse fl-chain. In addition, the nucleotide-derived sequences of the complete chain confirms the lack of carbohydrate, but reveals also that the mouse fl-chain is only three amino acids shorter than the human chain (642 and 645 amino acids) (Lundwall et al., 1984; De Bruijn and Fey, 1985). Second, swine C3b/iC3b/C3c generated by trypsin digestion of purified C3 or by storage of serum, have electrophoretic mobilities very close to that of intact C3. Consequently, and in contrast to the human condition (Bokisch et al., 1969), swine C3 fragments cannot easily be detected by crossed immunoelectrophoresis and used as indicators of complement activation. This may explain the failure to demonstrate C3 split products in synovial fluid of arthritic swine by immunoelectrophoresis (Timoney, 1976). On the other
PURIFICATION AND CHARACTERIZATION OF PORCINE C3
59
hand, the electrophoretic properties of C3dg in aged serum and oftryptic C3d are similar to the human counterparts (Lachmann et al., 1982), and in contrast to C3b/iC3b/C3c, the C3d fragment seems equally well-suited as a marker for the activation of porcine as well as human complement (Nilsson et al., 1990). This investigation has improved the separation techniques for swine C3 and indicated possible preparative procedures for the isolation of C4 and C5 of the same species. Knowledge has been extended regarding the characteristics of some activation products of swine C3 and a previously unrecognized difference in molecular weight between human and swine C3 fl-chain has been pointed out. Also, an antigenic cross-reaction between human and swine C3 is shown that could not be demonstrated by earlier preparations or reagents, and the herein reported amino acid composition correlates more strongly with human C3 than with earlier swine C3 determinations. ACKNOWLEDGEMENTS
We are deeply grateful to the late Professor Bo Ghane, Department of Animal Breeding and Genetics, SLU, Uppsala, who generously shared his great knowledge with us, and kindly performed the electrophoretic study of neuraminidase-treated C3. We are also thankful to Per-Erik Jansson, Department of Organic Chemistry, Arrhenius Laboratory, University of Stockholm, for kindly performing the carbohydrate analysis, Bo Nilsson for critical reading of the manuscript, Margita Nilsson and Susanne Heller for skillful technical assistance. This work was supported by grants from the Swedish Medical Research Council (no. 5674), the Swedish National Defence Research Institute, and from the Swedish National Board of Technical Development.
REFERENCES Barta, O. and Hubbert, N.L., 1978. Testing of hemolytic complement components in domestic animals. Am. J. Vet. Res., 39: 1303-1308. Bokisch, V.A., Dierich, M.P. and Mtiller-Eberhard, H.J., 1975. Third component of complement (C3): structural properties in relation to functions. Proc. Natl. Acad. Sci. USA, 72: 1989-1993. Bokisch, V.A., Miiller-Eberhard, H.J. and Cochrane, C.G., 1969. Isolation of a fragment (C3a) of the third component of human complement containing anaphylatoxin and chemotactic activity and description of an anaphylatoxin inactivator of human serum. J. Exp. Med., 129: 1109-1130. Borg, T., Alvfors, A., Gerdin, B. and Modig, J., 1985. A porcine model of early adult respiratory distress syndrome induced by endotoxemia. Acta Anaesthesiol. Scand., 29:814-830. Borsos, T., Dourmaskin, R.R. and Humphrey, J.H., 1964. Lesion in erythrocyte membranes caused by immune hemolysis. Nature, 202:251-252.
60
K.-E. S T O R M ET AL.
Cooper, N.R., 1988. Laboratory investigations of complement proteins and complement receptors. In: M.D. Kazatchkine and T. McMonagle (Editors), Complement and Immunological Disease. 2. Bailli6re Tindall, London, pp. 263-293. De Bruijn, M.H.L. and Fey, G.H., 1985. Human complement component C3: cDNA coding sequence and derived primary structure. Proc. Natl. Acad. Sci. USA, 82:708-712. Eggertsen, G., Hellman, U., Lundwall, ]~., Folkersen, J. and Sj6quist, J., 1985. Characterization of tryptic fragments of human complement factor C3. Mol. Immunol., 22:833-841. Hammer, C.H., Wirtz, G.H., Renfer, L., Gresham, H.D. and Tack, B.F., 1981. Large scale isolation of functionally active components of the human complement system. J. Biol. Chem., 256: (8): 3995-4006. Harboe, N.M.G. and Ingild, A., 1983. Immunization, isolation of immunoglobulins and antibody titre determination. In: N.H. Axelsen (Editor), Handbook of Immunoprecipitationin-Gel Techniques. Scandinavian Journal of Immunology. Supplement 10, Vol. 17. Blackwell, Oxford, UK, pp. 345-351. Kjaerwig Broe, M. and Ingild, A., 1983. Amplification ofimmunoprecipitates in agarose gels by horseradish peroxidase-labeled antibody. In: N.H. Axelsen (Editor), Handbook of Immunoprecipitation-in-Gel Techniques. Scandinavian Journal of Immunology. Supplement 10, Vol. 17. Blackwell, pp. 255-258. Lachmann, P.J., Pangburn, M.K. and Oldroyd, R.G., 1982. Breakdown of C3 after complement activation: identification of a new fragment, C3g, using monoclonal antibodies. J. Exp. Med., 156: 205-216. Laemmli, U.K. and Favre, M., 1973. Maturation of the head of bacteriophage T4. J. Mol. Biol., 80: 575-599. Lambris, J.D., 1988. The multifunctional role of C3, the third component of complement. Immunol. Today, 9( 12): 387-393. Linscott, W.D., 1986. Biochemistry and biology of the complement system in domestic animals. Prog. Vet. Microbiol. Immunol., 2: 54-77. Lundwall, ]~., Wetsel, R.A., Domdey, H., Tack, B.F. and Fey, G.H., 1984. Structure of murine complement component C3. I. Nucleotide sequence of cloned complementary and genomic DNA coding for the beta chain. J. Biol. Chem., 259:13851-13856. Mandle, R.J., McConnel-Mapes, J.A. and Nilsson, U.R., 1977. Inhibition of complement by mouse serum: selective inactivation of the fourth component of human complement. J. Immunol., 119(1): 180-186. Moore, S., 1963. On the determination of cystine as cysteic acid. J. Biol. Chem., 238( 1 ): 235237. Miiller-Eberhard, H.J., 1988. Molecular organization and function of the complement system. Annu. Rev. Biochem., 57: 321-347. Nilsson, B. and Nilsson, U.R., 1985. An assessment of the extent of antigenic analogy between physiologically bound C3 and C3 denatured by sodium dodecyl sulphate. Scand. J. Immunol., 22(6): 703-710. Nilsson, U.R., Miller, M.E. and Wyman, S., 1974. A functional abnormality of the fifth component of complement (C5) from human serum of individuals with a familial opsonic defect. J. Immunol., 112(3): 1164-1176. Nilsson, L., Storm, K.-E., Thelin, S., Bagge, L., Hultman, J., Thorelius, J. and Nilsson, U., 1990. Heparin-coated equipment reduces complement activation during cardiopulmonary bypass in the pig. Artif. Organs, 14( 1 ): 46-48. Nilsson, U.R., Tomar, R.H. and Taylor, F.B., 1972. Additional studies on human C5: development of a modified purification method and characterization of the purified product by polyacrylamide gel electrophoresis. Immunochemistry, 9: 709-723. Nonaka, M., Natsuume-Sakai, S., Kaidoh, T., Nonaka, M. and Takahashi, M., 1980. Characterization of allotypic determinants of murine complement C3. J. Immunol., 125: 2025-2030.
PURIFICATION AND CHARACTERIZATION OF PORCINE C3
61
Pangburn, M.K. and Mtiller-Eberhard, H.J., 1980. Relation of a putative thioesterbond in C3b to activation of the alternative pathway and the binding of C3b to biological targets of complement. J. Exp. Med., 152:1102-I 114. Pfiques, E.P., 1980. Purification and partial characterization of the third component of the complement system from porcine serum (C3) and of a crystallizable degradation product of the fourth component of the complement system from human serum (C4). Hoppe-Seyler's Z. Physiol. Chem., 361: 445-456. Sawardeker, J.S., Sloneker, J.H. and Jeanes, A., 1965. Quantitative determination of monosaccharides as their alditol acetates by gas liquid chromatography. Anal. Chem., 37: (12) 16021604. Spackman, D.H., Stein, W.H. and Moore, S., 1958. Automatic recording apparatus for use in the chromatography of amino acids. Anal. Chem., 30 (7): 1190-1206. Sundsmo, J.S. and Fair, D.S., 1985. Relationships among the complement, kinin, coagulation, and fibrinolytic systems. In: H.J. Miiller-Eberhard and P.A. Miescher (Editors), Complement. Springer, Berlin, pp. 379-406. Thomas, M.L. and Tack, B.F., 1983. Identification and alignment of a thiol ester site in the third component of guinea pig complement. Biochemistry, 22 (4): 942-947. Timoney, J.F., 1976. Erysipelas arthritis in swine: concentrations of complement and third component of complement in synovia. Am. J. Vet. Res., 37( 1 ): 5-8. Tomana, M., Niemann, M., Garner, C. and Volanakis, J.E., 1985. Carbohydrate composition of the second, third and fifth components and factors B and D of human complement. Mol. lmmunol., 22:107-111. Wallenborg, B. and Andersson, U.-B., 1978. Immunoelectrophoretic Techniques with the LKB 2117 Multiphor. Application Note 249, LKB-Produkter, Bromma, Sweden, 12 pp. Zidkov~, J., V~tvi~ka, V., Rossman, P., Dlaba~, V. and Strfinsky, J., 1984. Characterization of the third component of pig complement and its utilization in a C3b receptor study. Experientia, 40: 975-977.
